Enhanced surveillance involved the collection of a minimum dataset for each possible case, including demographic data, clinical symptoms, travel and contact history, and results of testing for respiratory pathogens (6). Nose and throat swab specimens and, when possible, lower respiratory tract specimens, were tested at 1 of 4 regional laboratories. Although the testing guidelines recommended MERS-CoV testing after exclusion of alternative etiologies, other tests were conducted in parallel with MERS-CoV testing for most suspected cases.

During the first few days of surveillance, a pan-coronavirus assay conducted at the PHE National Reference Laboratory was used as a screening test; then the viral genome was fully sequenced. After the generation of MERS-CoV specific assays, a first-line screening assay targeting the viral genomic area upstream of the E gene (7) was conducted, followed by confirmatory testing at the HPA/PHE National Reference Laboratory. Results of MERS-CoV testing are reported regularly in the HPA/PHE Weekly Influenza Report (8).

A descriptive analysis included the number of persons tested and proportion positive for MERS-CoV by key demographic, epidemiologic, and clinical characteristics. The positive predictive value of different combinations of signs and symptoms was calculated as the proportion of persons who had signs and symptoms of MERS and tested positive for MERS-CoV and showed exact (Clopper-Pearson) binomial 95% confidence intervals.

During September 24, 2012–October 15, 2013, 77 travelers from the Middle East that met the possible case definition were tested for MERS-CoV. Seventy-five travelers tested negative on the screening assay, and 2 tested positive. Positive results on the screening assay were confirmed by positive results at the HPA/PHE National Reference Laboratory.

In addition to testing the 77 persons who met all of the possible case criteria, MERS-CoV testing was conducted on 13 patients who had severe acute respiratory disease but did not meet the travel requirements: 2 had a travel history outside the Middle East, 4 had no travel history in the relevant exposure period, and travel histories of the remaining 7 were unknown. MERS-CoV was not detected in any of these persons.

The clinical and epidemiologic characteristics of the 77 persons tested and their MERS-CoV test results are shown in Table 1. Those tested ranged in age from 3 months to 90 years; 34 (44%) had signs of pulmonary parenchymal involvement. The 2 confirmed cases were in male patients, 45 and 60 years of age; both had severe acute respiratory symptoms requiring treatment by extracorporeal membrane oxygenation; both subsequently died. MERS-CoV PCR testing was conducted on 53 contacts of the 2 confirmed case-patients in England; 2 of these contacts tested MERS-CoV positive (9,10).

The positive predictive value for MERS-CoV infection of different combinations of signs and symptoms is shown in Table 2. No case-patients who did not have pulmonary parenchymal involvement tested positive for MERS-CoV, and the positive predictive value of the clinical manifestations increased as the severity of disease increased. Of the 77 patients tested, 22 had positive results for alternative respiratory pathogens, including 10 with influenza (7 influenza A and 3 influenza B); 1 of the influenza A–infected case-patients was later confirmed to also be infected with MERS-CoV. Two case-patients tested positive for Legionella pneumophila, 4 for rhinovirus, 3 for adenovirus, 1 for respiratory syncytial virus, and 1 for human metapneumovirus.

Unlike surveillance for established organisms, surveillance for a novel pathogen requires analysis of information collected from all patients tested, even from those that test negative, to build knowledge of the predictive value of different epidemiologic and clinical manifestations. This report on the characteristics of patients traveling to England from the Middle East and tested for MERS-CoV enables a first crude estimation of the positive predictive value of different signs and symptoms during the first year following the emergence of this pathogen.

Because this study is based on a cohort of 77 suspected case-patients, of whom only 2 laboratory-confirmed cases were identified during the surveillance period, estimates on the basis of identified symptoms are very imprecise (Table 2). However, in the context of an emerging pathogen, reporting such data progressively helps optimize case detection and surveillance systems.

During the 12-month surveillance period, no patients who had respiratory symptoms but no pulmonary parenchymal involvement were positive for MERS-CoV by PCR, and the positive predictive value of signs and symptoms increased with the severity of clinical manifestation. This suggests that the case definitions that were in use during this period (which recommended MERS-CoV testing only for patients who met the epidemiologic criteria and had a severe respiratory illness) were appropriate.

A range of respiratory pathogens were found in those patients that were MERS-CoV negative, highlighting the importance of looking for alternative diagnoses. However, the diagnosis of 1 of the MERS-CoV case-patients was delayed because of an initial diagnosis of influenza. The testing algorithm was subsequently changed to ensure that patients meeting the possible case definition were tested for MERS-CoV if they had an alternative etiology which did not fully explain their clinical manifestation.

The predictive value of the possible case definition depends on the incidence of infection and would be expected to vary across different population groups and change over time, especially in the context of an emerging pathogen. We encourage other countries to similarly report the characteristics of all patients tested for MERS-CoV to improve understanding of the predictive value of different clinical and epidemiologic manifestations in various populations at different times. This will help inform the evolving international public health response to this novel pathogen.